Back to vendors
I

Imbue

Also known as: Imbue Sculptor, Generally Intelligent

Visit site
Entry priceFree and open source (MIT); uses your own Claude planFull pricing detail

Desktop app from Imbue, an AI research lab working on verifying the correctness of AI generated code, for running parallel coding agents, each in its own isolated git worktree, with every change reviewed by the developer before it is merged.

Imbue is a San Francisco AI research lab, formerly known as Generally Intelligent and founded in 2018, that builds tools to make AI coding reliable, collaborative, and accessible. Led by chief executive Kanjun Qiu, it has raised roughly two hundred thirty two million dollars from investors including NVIDIA, Eric Schmidt, Cruise founder Kyle Vogt, and Notion co founder Simon Last, and it operates a ten thousand graphics processor research cluster. Its research mission is to improve the ability to verify the correctness of code generated by large language models, so people can trust the software they build.

Imbue's first product is Sculptor, a desktop application for macOS and Linux for running parallel coding agents. Each agent works in its own isolated git worktree with its own branch and terminal, with a Docker or remote container backend available as an experimental option, and five or more agents can run at once from one workspace. Sculptor offers Claude models out of the box on the developer's own Claude plan, and its support for the Pi agent harness lets developers choose any provider or run any terminal based agent. It is designed to let a developer kick off a refactor with one agent while another builds a feature and a third checks a hunch.

Control sits at the center. Each agent's changes reach the codebase only when the developer reviews the diff and merges, Pairing Mode syncs an agent's work into the local editor so changes can be tested and reviewed, and Sculptor opens GitHub pull requests and tracks their status. A built in library of workflow skills runs as full agents, including a fix bug skill that writes failing tests before fixing, and a CI Babysitter dispatches agents to fix failing pipelines. Every session is saved with its plans, chats, tool calls, and code changes intact, and sessions can be resumed or forked from any point.

Sculptor runs fully locally, with granular control over what data returns to Imbue and no training on a user's code by default. It is free and open source under the MIT license, and the customer pays the model provider directly. As a focused developer tool rather than a general business agent platform, it does not offer enterprise security certifications or business system integrations. Its strength is orchestrating coding agents while keeping every change under the developer's review.

Vendor details

Canonical URL

https://imbue.com

Category

Agent builder

Subcategory

Desktop harness for running and verifying parallel coding agents in isolated containers

Funding status

Independent research lab. Founded 2018 as Generally Intelligent by Kanjun Qiu and Josh Albrecht, renamed Imbue, based in San Francisco. Reported to have raised roughly $212M to $232M from investors including NVIDIA, Eric Schmidt, Kyle Vogt and Simon Last, operating a large research GPU cluster. Sculptor is its first product, in beta and free during beta, with the customer supplying their own Anthropic API key or Claude Pro or Max subscription; long-term pricing is not published. Public repository last active May 2026 at v0.30.

Company status

independent

Use cases & customers

Primary use cases

Running several coding agents in parallel, each in an isolated container, without dependency collisions or risk to the local machineSyncing any agent's work into the local repository and IDE to test it before mergingEnforcing plain-English rules on agent output through instruction audits, and catching tests that falsely report passingUsing Claude Code or Codex with your own API key or Claude subscription, with no vendor between you and the modelKeeping source code entirely on the developer's own machine, with granular control over what telemetry returns to the vendor

Deployment options

local

Integrations

Sculptor integrates with the developer's own toolchain rather than business systems: local git repositories, with each agent in its own worktree and branch and Pairing Mode syncing an agent's work into the local repository, the developer's own IDE and dev environment, an experimental Docker or remote container backend, devcontainer configuration with port forwarding, and linters including pylint, ruff and pyre integrated into verification. Claude Code runs on the customer's own Claude plan, and the Pi agent harness opens any provider and any terminal-based agent. Customers can build their own Sculptor plugins in JavaScript or TypeScript and swap the agent harness entirely, and Sculptor opens GitHub pull requests and tracks their status. No business system connectivity exists and none is intended; tool calling within a task belongs to the underlying coding agents.

In practice

A developer wants to explore three approaches at once. In Sculptor they spin up three agents, each in its own container, kicking off a refactor, a new feature, and a bug fix in parallel, then jump between them without environment conflicts.

An engineer worries an agent faked a passing test. Sculptor's Suggestions flag the misleading behavior, and an instruction audit rule written in plain English catches other violations before the code is merged into the local repository.

A developer returns to a project after a week. They reopen the saved Sculptor session with its plans, chats, tool calls, and code changes intact, and pick up exactly where the agent left off without re prompting.

Agentic Index coverage score

10.0 / 14 capabilities · 71%

Integrations & Tool Calling Partial

Sculptor integrates with local git repositories, keeping files and git state synced between an agent's container and the developer's local repository through Pairing Mode, and works alongside the developer's own IDE and development environment. Docker provides per-agent containers with full repository copies, with devcontainer port forwarding added in a 2026 release.

Linters including pylint, ruff and pyre are integrated into the verification checks. Supported coding agents are Claude Code and Codex, with the vendor inviting requests for others. No business system connectivity such as CRM, ticketing, messaging or data warehouse integration is documented, and tool calling within a task is performed by the underlying coding agents rather than by Sculptor.

Sourceimbue.com sculptor product page and Sculptor announcement, public release notesread 2026-08-31

Workflow Orchestration Full

The vendor documents running five or more coding agents in parallel, each in an isolated worktree with its own branch, terminal and diff view, all visible and mergeable from one workspace, and skills that run as full agents able to spawn parallel subagents. A developer reviews and merges each agent's changes, and Sculptor opens GitHub pull requests and tracks their status. Multiple agents with subagent spawning.

Sourceimbue.com/product/sculptor and github.com/imbue-ai/sculptor READMEread 2026-09-29

Knowledge Grounding & RAG Partial

Each coding agent runs in a container with a full copy of the developer's repository, so grounding is in the complete codebase rather than retrieved fragments, with devcontainer support providing the project's environment configuration. Instruction audit rules written by the developer in plain English persist as policy applied to agent work.

No index, ingestion pipeline, document or wiki knowledge source, curated knowledge object or retrieval configuration maintained by Sculptor over and above the repository appears in the product documentation; the repository is maintained by the customer through git and copied into the container.

Sourceimbue.com sculptor product page and Sculptor announcementread 2026-08-31

Human Oversight & Guardrails Partial

Each Sculptor agent works in its own worktree and branch, pushes to a local git remote, and its changes reach the codebase only when the developer reviews the diff and merges, which the vendor describes as full review before merge. That is a customer merge gate rather than a vendor approval step inside the agent's run. Instruction audits, plain English rules checked against an agent's work, were listed under What's next in the September 2025 announcement and do not appear on the current product page or README.

Sourceimbue.com/product/sculptor, Sculptor announcement and github.com/imbue-ai/sculptor READMEread 2026-09-29

Security, Identity & Governance Partial

Each coding agent runs in its own Docker container so it can install packages and execute code without risking the developer's machine, which the vendor contrasts with git worktrees that share the local environment.

Sculptor runs fully locally with documented levels controlling what returns to Imbue, the Essential-only level sharing just error logs and basic usage and described by the vendor as the safe choice for work code, and code is not used for training by default.

No security certification, trust center, single sign-on, role-based access control or team-level audit capability is documented; the product is a single-developer desktop application in beta.

Sourceimbue.com sculptor product page and Sculptor announcementread 2026-08-31

Observability & Auditability Full

The vendor states every agent session is saved with its plans, chats, tool calls and code changes intact, and that sessions can be resumed and forked from any point. Suggestions flags misleading agent behavior, with the vendor citing tests that falsely report passing as an example.

Pairing Mode brings an agent's work from its container into the local repository with files and git state synced, making changes inspectable in the developer's own editor and diff tooling, and merges surface conflicts for resolution. No aggregate cross-session analytics, fleet reporting or organizational dashboard is documented.

Sourceimbue.com Sculptor announcement and sculptor product pageread 2026-08-31

Memory & State Persistence Partial

The vendor states every agent session is saved with its plans, chats, tool calls and code changes intact, that sessions can be resumed, and that agents can be forked from any point, so context carries across sessions. Instruction audit rules are written and maintained by the developer. No capability is documented for an agent to learn from corrections, carry conclusions between separate sessions, or update its own instructions from experience, and agents are deliberately isolated from one another in separate containers.

Sourceimbue.com Sculptor announcement and sculptor product pageread 2026-08-31

Deployment & Data Residency Full

The vendor states Sculptor runs on the customer's own computer rather than Imbue's servers, that data never leaves the machine and Imbue does not store repositories or train on code, and that it is free, local and MIT-licensed open source. Each agent works in an isolated git worktree with its own branch and terminal, with a Docker or remote container backend documented as experimental. Distribution covers macOS on Apple Silicon and Linux including ARM64.

Sourceimbue.com/product/sculptor and github.com/imbue-ai/sculptor READMEread 2026-09-29

Prebuilt Agents, Templates & Packs Full

Sculptor ships a built-in library of workflow skills that run as full agents with their own tools: spec writing, mock generation and a fix-bug skill that runs a reproduction interview, writes failing tests and then makes them pass, with experimental skills documented separately and a CI Babysitter that fixes failing pipelines. That is a small named set of ready agents shipped with the product, and the set is short.

Sourcegithub.com/imbue-ai/sculptor README and imbue.com/product/sculptorread 2026-09-29

Triggers & Channel Coverage Full

The vendor's Sculptor product page states the CI Babysitter automatically dispatches agents to fix failing pipelines, an autonomous wake on a CI event, and Sculptor opens GitHub pull requests and tracks their status. Otherwise a developer launches agents on demand from the desktop app on macOS or Linux. The CI Babysitter rests on one FAQ line and does not appear in the README's feature summary.

Sourceimbue.com/product/sculptor and github.com/imbue-ai/sculptor READMEread 2026-09-29

Model Flexibility & Routing Full

Sculptor offers Claude models out of the box (Opus 4.8, 4.7 and 4.6, Sonnet 4.6, Haiku 4.5) on the customer's own Claude plan, and ships integrated support for the Pi agent harness, through which the customer chooses any provider or switches models mid-session; it can also run any terminal-based agent. That is customer choice across model makers. The product page's FAQ still calls Pi support experimental, and Codex is no longer listed.

Sourceimbue.com/product/sculptor and github.com/imbue-ai/sculptor READMEread 2026-09-29

APIs, SDKs & MCP Extensibility Full

The vendor's product page states customers can build their own Sculptor plugins in JavaScript or TypeScript and swap the agent harness entirely, and the README states Sculptor ships integrated support for the Pi harness alongside Claude Code and can run and manage any terminal-based agent. Sculptor is open source under the MIT license with a public repository and documentation. The plugin interface is stated on the product page, and a plugin reference has not yet been recorded.

Sourceimbue.com/product/sculptor and github.com/imbue-ai/sculptor READMEread 2026-09-29

Testing, Debugging & Optimization Partial

The September 2025 Sculptor announcement lists Suggestions as a beta that points out fixes to agent code, and places instruction audits and flagging misleading behavior such as tests that pass without real tests under What's next, so those were roadmap features.

The current product page and README describe per-agent diff review, a fix-bug skill that writes failing tests before fixing, and a CI Babysitter, and neither mentions Suggestions or audits. Tests the agent writes and runs are the agent's method rather than a harness scoring the agent's output. The vendor's separate open-source tool for catching agent mistakes is another product.

Sourceimbue.com Sculptor announcement, imbue.com/product/sculptor and github.com/imbue-ai/sculptor READMEread 2026-09-29

Browser & Computer Use Not documented

Coding agents execute code, install dependencies and run terminals and test suites inside isolated Docker containers, which is code execution rather than operation of software lacking a programmatic interface.

Devcontainer port forwarding allows a developer to view an application running inside a container from their own browser, which is a human inspecting the agent's output. All agent action reaches the filesystem, git, package managers and linters programmatically within the container boundary. No browser control, navigation, form filling, screen interaction or desktop automation appears in the product documentation.

Sourceimbue.com sculptor product page and Sculptor announcement, public release notesread 2026-08-31

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Pricing

Free and open source (MIT); uses your own Claude plan

Agentic Index verified 2026-09-29

Alternatives to Imbue

The closest documented capability profiles to Imbue among agent builders tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

  • AgentX12.0 / 14Fuller documented coverage on Integrations & Tool Calling and Knowledge Grounding & RAG
  • Airia12.0 / 14Fuller documented coverage on Integrations & Tool Calling and Knowledge Grounding & RAG
  • Airtable11.0 / 14Fuller documented coverage on Integrations & Tool Calling and Knowledge Grounding & RAG
  • Dify12.0 / 14Fuller documented coverage on Integrations & Tool Calling and Knowledge Grounding & RAG
  • Langflow12.0 / 14Fuller documented coverage on Integrations & Tool Calling and Knowledge Grounding & RAG
  • Make11.0 / 14Fuller documented coverage on Integrations & Tool Calling and Knowledge Grounding & RAG

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Head to head

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.